The long term objective of this project is to understand the molecular details of antigen recognition by T-cells, and to use this knowledge for the rational design of a safe, highly immunogenic, carrier free vaccine for inducing T and B cell immunity to the AIDS virus. Unlike B cells which form a binary complex with native protein in solution, T-cells recognize antigen which has been proteolytically digested or otherwise processed, and then only on the surface of another cell--usually a B-cell, macrophage or dendritic cell. Presentation occurs in conjunction with a class I (for T8 cells) or class II (for T4 cells) product of the major histocompatibility gene complex. Understanding the molecular details of recognition thus requires elucidating the structure of a ternary complex: the T cell antigen receptor, the antigen fragment, and the MHC product. Our approach to these problems will involve computation and experimental testing. The key to the computational problem is the close homology between immunoglobulins, T-cell receptors, and the class I and class II MHC products. The crystal structures for a number of immunoglobulins have been determined, as has the structure of a class I molecule. Enough information is available so that significant homologies can be found between all domains of unknown structure and one or more domains of known structure of a molecule from the known structure of a homologous molecules. These methods will be further developed where necessary and applied to the structural determination of the domains of interest. Objectives of general interest are finding the location and structure of the antigen binding site on the class I and class II MHC products, finding the locations and structure of the antigen binding site on T-cell receptors, and locating the sites of interactions between the MHC product and T-cell receptor. More specific goal are related to predicting the effect of a change in any specified residue in any of the three components, on he stability of the complex and on antigenicity. These will be tested experimentally and used to guide site specified mutagenesis experiments, and the identification of portions of the AIDS virus gp120 protein that are immunogenic for helper and cytotoxic T cells in mice.